A naturally produced antibody (Ab) is a tetrameric structure consisting of two identical immunoglobulin (Ig) heavy chains and two identical light chains. The heavy and light chains of an Ab consist of different domains. Each light chain has one variable domain (VL) and one constant domain (CL), while each heavy chain has one variable domain (VH) and three or four constant domains (CH). Each domain, consisting of about 110 amino acid residues, is folded into a characteristic β-sandwich structure formed from two β-sheets packed against each other, the immunoglobulin fold. The VL domains each have three complementarity determining regions (CDR1-3) and the VH domains each have up to four complimentarity determining regions (CDR1-4), that are loops, or turns, connecting β-strands at one end of the domains. The variable regions of both the light and heavy chains generally contribute to antigen specificity, although the contribution of the individual chains to specificity is not necessarily equal. Antibody molecules have evolved to bind to a large number of molecules by using randomized CDR loops.
Functional substructures of Abs can be prepared by proteolysis and by recombinant methods. They include the Fab fragment, which comprises the VH-CH1 domains of the heavy chain and the VL-CL1 domains of the light chain joined by a single interchain disulfide bond, and the Fv fragment, which comprises only the VH and VL domains, and the Fc portion which comprises the non-antigen binding region of the molecule. In some cases, a single VH domain retains significant affinity for antigen (Ward et al., 1989, Nature 341, 554-546). It has also been shown that a certain monomeric K light chain will specifically bind to its antigen. (L. Masat et al., 1994, PNAS 91:893-896). Separated light or heavy chains have sometimes been found to retain some antigen-binding activity as well (Ward et al., 1989, Nature 341, 554-546).
Another functional substructure is a single chain Fv (scFv), comprised of the variable regions of the immunoglobulin heavy and light chain, covalently connected by a peptide linker (S-z Hu et al., 1996, Cancer Research, 56, 3055-3061). These small (Mr 25,000) proteins generally retain specificity and affinity for antigen in a single polypeptide and can provide a convenient building block for larger, antigen-specific molecules. The short half-life of scFvs in the circulation limits their therapeutic utility in many cases.
A small protein scaffold called a “minibody” was designed using a part of the Ig VH domain as the template (Pessi et al., 1993, Nature 362, 367-369). Minibodies with high affinity (dissociation constant (Kd) about 10−7 M) to interleukin-6 were identified by randomizing loops corresponding to CDR1 and CDR2 of VH and then selecting mutants using the phage display method (Martin et al., 1994, EMBO J. 13, 5303-5309).
Camels often lack variable light chain domains when IgG-like material from their serum is analyzed, suggesting that sufficient antibody specificity and affinity can be derived from VH domains (three or four CDR loops) alone. “Camelized” VH domains with high affinity have been made, and high specificity can be generated by randomizing only the CDR3.
An alternative to the “minibody” is the “diabody.” Diabodies are small bivalent and bispecific antibody fragments, having two antigen-binding sites. The fragments comprise a heavy-chain variable domain (VH) connected to a light-chain variable domain (VL) on the same polypeptide chain (VH-VL). Diabodies are similar in size to the Fab fragment. By using a linker that is too short to allow pairing between the two domains on the same chain, the domains are forced to pair with the complementary domains of another chain and create two antigen-binding sites. These dimeric antibody fragments, or “diabodies,” are bivalent and bispecific. See, P. Holliger et al., PNAS 90:6444-6448 (1993).
CDR peptides and organic CDR mimetics have been made (Dougall et al., 1994, Trends Biotechnol. 12, 372-379). CDR peptides are short, typically cyclic, peptides which correspond to the amino acid sequences of CDR loops of antibodies. CDR loops are responsible for antibody-antigen interactions. CDR peptides and organic CDR mimetics have been shown to retain some binding affinity (Smyth & von Itzstein, 1994, J. Am. Chem. Soc. 116, 2725-2733). Mouse CDRs have been grafted onto the human Ig framework without the loss of affinity (Jones et al., 1986, Nature 321, 522-525; Riechmann et al., 1988).
In the body, specific Abs are selected and amplified from a large library (affinity maturation). The processes can be reproduced in vitro using combinatorial library technologies. The successful display of Ab fragments on the surface of bacteriophage has made it possible to generate and screen a vast number of CDR mutations (McCafferty et al., 1990, Nature 348, 552-554; Barbas et al., 1991, Proc. Natl. Acad. Sci. USA 88,7978-7982; Winter et al., 1994, Annu. Rev. Immunol. 12, 433-455). An increasing number of Fabs and Fvs (and their derivatives) are produced by this technique. The combinatorial technique can be combined with Ab mimics.
A number of protein domains that could potentially serve as protein scaffolds have been expressed as fusions with phage capsid proteins. Review in Clackson & Wells, Trends Biotechnol. 12:173-184 (1994). Several of these protein domains have already been used as scaffolds for displaying random peptide sequences, including bovine pancreatic trypsin inhibitor (Roberts et al., PNAS 89:2429-2433 (1992)), human growth hormone (Lowman et al., Biochemistry 30:10832-10838 (1991)), Venturini et al., Protein Peptide Letters 1:70-75 (1994)), and the IgG binding domain of Streptococcus (O'Neil et al., Techniques in Protein Chemistry V (Crabb, L., ed.) pp. 517-524, Academic Press, San Diego (1994)). These scaffolds have displayed a single randomized loop or region. Tendamistat has been used as a presentation scaffold on the filamentous phage M13 (McConnell and Hoess, 1995, J. Mol. Biol. 250:460-470).
Receptor tyrosine kinases of the ErbB family play pivotal roles in cell growth and differentiation. Aberrant activation of these receptors is associated with human cancers. Dimerization (the pairing of receptors) is essential to the signaling activity of all ErbB receptors. Blocking the dimerization activity of ErbB2 has been shown to directly inhibit the ability of ErbB2 to dimerize with other ErbB receptor proteins. Inhibiting receptor dimerization prevents the activation of ErbB signaling pathways. An antagonistic molecule that down regulates ErbB signaling could function as an anti-tumor agent. The ErbB signaling network is currently a major target in the development of anti-tumor drugs. ErbB-1 is a specific receptor for EGF, while ErbB-2 has no known natural ligand. ErbB2 is able to form heterodimers with ErbB-1 upon addition of EGF. ErbB2 also functions as the preferred dimerization partner for the kinase-dead ErbB-3 and for ErbB-4, which are both receptors for the neuregulins. The ErbB signaling network can also be activated in an indirect manner during signaling by cytokines and ligands of G-coupled protein receptors, indicating that it plays a central role in the growth control of many different cell types.
The proto-oncogene c-erbB-1 encodes the epidermal growth factor receptor. Its name originates from the viral homolog v-erbB which was isolated from an avian erythroblastosis virus (AEV) where it was contained as a fragment of the chicken c-ErbB-1 gene lacking the amino-terminal ligand-binding domain. Over expression of erbB-1 genes occurs in a wide range of tumors, including squamous carcinomas of various sites and adenocarcinomas. The human c-erbB-1 gene is located in the chromosomal region 7p14 and 7p12.
The ErbB-2 proto-oncogene (also referred to as Neu, EGFR-2 or HER-2) is a member of the transmembrane receptor tyrosine kinase family, which also includes EGF receptor and EGFR-3 (HER-3 or ErbB-3). ErbB-2 encodes a transmembrane receptor-like glycoprotein of 185 kDa with intrinsic tyrosine kinase activity. Although, ErbB-2 does not have any known high-affinity ligands, its kinase activity can be activated without ligand by either over expression or hetero-association with other members of the ErbB family of receptors. Amplification of the ErbB-2 gene and over expression of its product has been detected in almost 40% of primary human breast tumors. ErbB-2 over expression is also observed in ovarian, gastric, salivary and non-small cell lung carcinomas. ErbB-2 is activated by the neuregulins in heterodimers with the neuregulin receptors ErbB-3 and ErbB-4. The humanized anti-ErbB-2 monoclonal antibody Herceptin (from monoclonal 4D5) has received FDA approval for treatment of cancers that over express ErbB-2. Another anti-ErbB2 antibody in development is Pertuzumab (from monoclonal 2C4). Specific inhibitors of the tyrosine kinase activity of ErbB-1 (EGF receptor) are also in clinical trials.
Anti-ErbB2 antibodies are known in the art, and include but are not limited to U.S. Pat. Nos. 4,753,894; 5,169,774; 5,677,171; 5,720,937; 5,720,954; 5,725,856; 5,770,195; 5,772,997; 5,783,186; 6,054,561; 6,165,464; 6,333,169; 6,015,567; 6,387,371; 6,399,063; 6,441,143; 6,458,356; 6,627,196, each of which is incorporated by reference herein.
Covalent attachment of the hydrophilic polymer poly(ethylene glycol), abbreviated PEG, is a method of increasing water solubility, bioavailability, increasing serum half-life, increasing therapeutic half-life, modulating immunogenicity, modulating biological activity, or extending the circulation time of many biologically active molecules, including proteins, peptides, and particularly hydrophobic molecules. PEG has been used extensively in pharmaceuticals, on artificial implants, and in other applications where biocompatibility, lack of toxicity, and lack of immunogenicity are of importance. In order to maximize the desired properties of PEG, the total molecular weight and hydration state of the PEG polymer or polymers attached to the biologically active molecule must be sufficiently high to impart the advantageous characteristics typically associated with PEG polymer attachment, such as increased water solubility and circulating half life, while not adversely impacting the bioactivity of the parent molecule.
PEG derivatives are frequently linked to biologically active molecules through reactive chemical functionalities, such as lysine, cysteine and histidine residues, the N-terminus and carbohydrate moieties. Proteins and other molecules often have a limited number of reactive sites available for polymer attachment. Often, the sites most suitable for modification via polymer attachment play a significant role in receptor binding, and are necessary for retention of the biological activity of the molecule. As a result, indiscriminate attachment of polymer chains to such reactive sites on a biologically active molecule often leads to a significant reduction or even total loss of biological activity of the polymer-modified molecule. R. Clark et al., (1996), J. Biol. Chem., 271:21969-21977. To form conjugates having sufficient polymer molecular weight for imparting the desired advantages to a target molecule, prior art approaches have typically involved random attachment of numerous polymer arms to the molecule, thereby increasing the risk of a reduction or even total loss in bioactivity of the parent molecule.
Reactive sites that form the loci for attachment of PEG derivatives to proteins are dictated by the protein's structure. Proteins, including enzymes, are composed of various sequences of alpha-amino acids, which have the general structure H2N—CHR—COOH. The alpha amino moiety (H2N—) of one amino acid joins to the carboxyl moiety (—COOH) of an adjacent amino acid to form amide linkages, which can be represented as —(NH—CHR—CO)n—, where the subscript “n” can equal hundreds or thousands. The fragment represented by R can contain reactive sites for protein biological activity and for attachment of PEG derivatives.
For example, in the case of the amino acid lysine, there exists an —NH2 moiety in the epsilon position as well as in the alpha position. The epsilon —NH2 is free for reaction under conditions of basic pH. Much of the art in the field of protein derivatization with PEG has been directed to developing PEG derivatives for attachment to the epsilon —NH2 moiety of lysine residues present in proteins. “Polyethylene Glycol and Derivatives for Advanced PEGylation”, Nektar Molecular Engineering Catalog, 2003, pp. 1-17. These PEG derivatives all have the common limitation, however, that they cannot be installed selectively among the often numerous lysine residues present on the surfaces of proteins. This can be a significant limitation in instances where a lysine residue is important to protein activity, existing in an enzyme active site for example, or in cases where a lysine residue plays a role in mediating the interaction of the protein with other biological molecules, as in the case of receptor binding sites.
A second and equally important complication of existing methods for protein PEGylation is that the PEG derivatives can undergo undesired side reactions with residues other than those desired. Histidine contains a reactive imino moiety, represented structurally as —N(H)—, but many chemically reactive species that react with epsilon —NH2 can also react with —N(H)—. Similarly, the side chain of the amino acid cysteine bears a free sulfhydryl group, represented structurally as —SH. In some instances, the PEG derivatives directed at the epsilon —NH2 group of lysine also react with cysteine, histidine or other residues. This can create complex, heterogeneous mixtures of PEG-derivatized bioactive molecules and risks destroying the activity of the bioactive molecule being targeted. It would be desirable to develop PEG derivatives that permit a chemical functional group to be introduced at a single site within the protein that would then enable the selective coupling of one or more PEG polymers to the bioactive molecule at specific sites on the protein surface that are both well-defined and predictable.
In addition to lysine residues, considerable effort in the art has been directed toward the development of activated PEG reagents that target other amino acid side chains, including cysteine, histidine and the N-terminus. See, e.g., U.S. Pat. No. 6,610,281 which is incorporated by reference herein, and “Polyethylene Glycol and Derivatives for Advanced PEGylation”, Nektar Molecular Engineering Catalog, 2003, pp. 1-17. A cysteine residue can be introduced site-selectively into the structure of proteins using site-directed mutagenesis and other techniques known in the art, and the resulting free sulfhydryl moiety can be reacted with PEG derivatives that bear thiol-reactive functional groups. This approach is complicated, however, in that the introduction of a free sulfhydryl group can complicate the expression, folding and stability of the resulting protein. Thus, it would be desirable to have a means to introduce a chemical functional group into bioactive molecules that enables the selective coupling of one or more PEG polymers to the protein while simultaneously being compatible with (i.e., not engaging in undesired side reactions with) sulfhydryls and other chemical functional groups typically found in proteins.
As can be seen from a sampling of the art, many of these derivatives that have been developed for attachment to the side chains of proteins, in particular, the —NH2 moiety on the lysine amino acid side chain and the —SH moiety on the cysteine side chain, have proven problematic in their synthesis and use. Some form unstable linkages with the protein that are subject to hydrolysis and therefore decompose, degrade, or are otherwise unstable in aqueous environments, such as in the bloodstream. Some form more stable linkages, but are subject to hydrolysis before the linkage is formed, which means that the reactive group on the PEG derivative may be inactivated before the protein can be attached. Some are somewhat toxic and are therefore less suitable for use in vivo. Some are too slow to react to be practically useful. Some result in a loss of protein activity by attaching to sites responsible for the protein's activity. Some are not specific in the sites to which they will attach, which can also result in a loss of desirable activity and in a lack of reproducibility of results. In order to overcome the challenges associated with modifying proteins with poly(ethylene glycol) moieties, PEG derivatives have been developed that are more stable (e.g., U.S. Pat. No. 6,602,498, which is incorporated by reference herein) or that react selectively with thiol moieties on molecules and surfaces (e.g., U.S. Pat. No. 6,610,281, which is incorporated by reference herein). There is clearly a need in the art for PEG derivatives that are chemically inert in physiological environments until called upon to react selectively to form stable chemical bonds.
Recently, an entirely new technology in the protein sciences has been reported, which promises to overcome many of the limitations associated with site-specific modifications of proteins. Specifically, new components have been added to the protein biosynthetic machinery of the prokaryote Escherichia coli (E. coli) (e.g., L. Wang, et al., (2001), Science 292:498-500) and the eukaryote Sacchromyces cerevisiae (S. cerevisiae) (e.g., J. Chin et al., Science 301:964-7 (2003)), which has enabled the incorporation of non-genetically encoded amino acids to proteins in vivo. A number of new amino acids with novel chemical, physical or biological properties, including photoaffinity labels and photoisomerizable amino acids, keto amino acids, and glycosylated amino acids have been incorporated efficiently and with high fidelity into proteins in E. coli and in yeast in response to the amber codon, TAG, using this methodology. See, e.g., J. W. Chin et al., (2002), Journal of the American Chemical Society 124:9026-9027; J. W. Chin, & P. G. Schultz, (2002), ChemBioChem 11:1135-1137; J. W. Chin, et al., (2002), PNAS United States of America 99:11020-11024; and, L. Wang, & P. G. Schultz, (2002), Chem. Comm., 1-10. These studies have demonstrated that it is possible to selectively and routinely introduce chemical functional groups, such as ketone groups, alkyne groups and azide moieties, that are not found in proteins, that are chemically inert to all of the functional groups found in the 20 common, genetically-encoded amino acids and that may be used to react efficiently and selectively to form stable covalent linkages.
The ability to incorporate non-genetically encoded amino acids into proteins permits the introduction of chemical functional groups that could provide valuable alternatives to the naturally-occurring functional groups, such as the epsilon —NH2 of lysine, the sulfhydryl —SH of cysteine, the imino group of histidine, etc. Certain chemical functional groups are known to be inert to the functional groups found in the 20 common, genetically-encoded amino acids but react cleanly and efficiently to form stable linkages. Azide and acetylene groups, for example, are known in the art to undergo a Huisgen [3+2] cycloaddition reaction in aqueous conditions in the presence of a catalytic amount of copper. See, e.g., Tornoe, et al., (2002) Org. Chem. 67:3057-3064; and, Rostovtsev, et al., (2002) Angew. Chem. Int. Ed. 41:2596-2599. By introducing an azide moiety into a protein structure, for example, one is able to incorporate a functional group that is chemically inert to amines, sulfhydryls, carboxylic acids, hydroxyl groups found in proteins, but that also reacts smoothly and efficiently with an acetylene moiety to form a cycloaddition product. Importantly, in the absence of the acetylene moiety, the azide remains chemically inert and unreactive in the presence of other protein side chains and under physiological conditions.
The present invention addresses, among other things, problems associated with the activity and production of antigen-binding polypeptides and fragments thereof, and also addresses the production of antigen-binding polypeptides with improved biological or pharmacological properties, such as improved therapeutic half-life.